FIXME: Make it able to do for all images. Create a pmap that will convert .jpgs -> .png's
In the near future usage will be soon available on http://dc9.x10host.com/
Drag images you want to be prepared into resources/ See images into trained and test directories
(build-image-data-txt "nCov-training-set" "train-dir/" "test-dir/" "src/cov_rgb_bgr_converter/paths.txt" 52) ; with a txt file
(generated-txt "CT_COVID/" "src/nCov.txt")
(generated-txt "ants/" "src/path-ants.txt")
(resize-image-export "exports/" "ants/ant.png" 100 100)
(invert-image-export "exports/" "ants/ant.png")
(grayscale-resize "exports/" "ants/ant.png" 100)
(generated-txt "dog-cat/" "src/train-cat-dog.txt")
(build-folder "training-set-name" "train-dir" "output-dir")
(build-image-data-txt "training-set-name" "train-dir" "output-dir" "src/train-cat-dog.txt" 52)
See examples of .txt generated in src/.
- Will generate .txt file when you drag a folder into resources directory.
- Will take .png files only (convert file using magick?)
- (see Usage)
Steps: (will be implemented via commandline)
- Build project in directory using leinigen
- Drag folder of images into resources/
- call 'lein repl'
- call methods in core.clj
- Implement it into a command line BIG
- We want to automate the process we want to be able to drag in any type of data set (nCov, dog and cat pictures, etc. , (.txt data)) mkdir /test-nCov-test ... or future /test-[insert-data-set-name]
- RESTFUL API (https://www.tutsmake.com/node-express-js-creating-a-restful-api-mysql-example/)
- train the data (Fiverr look into it)
- Show results (Via web server -> create app that will talk to processing backend)
- take in .xlxs data spreadsheet or .txt
- Take in video files for interpolation (guess speed of vehicle - comma.ai challenge)
- .txt data (NLP)
- TBD.
- Parse through the image. DONE
- Grayscale the pixels. DONE
- Export it as a new photo or image. DONE
- mapping across a folder - convert all images to gray scale. DONE
- Preparing training data to look at. DONE
- Optional: create an autoupdate for paths.txt. - for instance, when user drags in a dataset into resources folder a paths.txt should be generated. DONE
- jpg's won't allow filters
- http://dc9.x10host.com/ (implement the functionality in the command line on my website)
- https://cljdoc.org/d/net.mikera/imagez/0.12.0/api/mikera.image.dither#greyscale-palette-function
- https://drive.google.com/drive/folders/1kBer7h7ljCbCHqcyUikH0fisPEFGDV0e?usp=sharing (COVID-19 images/resources)
- project will serve as a baseline to how to convert any image into grayscale. Will take in directories, check if image is a .png will do everything to prepare data for ML.
- Library here will serve to prepare all types of data in the future (will be updated)
- generalized parameters -> user only has to drag images library into resources folder
- create a script that will update paths.txt?
- prepare-example.clj (we want to prepare nCov-2019 images soon - will serve as an example)
- more COVID-CT images (have a libary of images stored offsite and in Google drive)
- Parse through the image DONE
- grayscale the pixels DONE
- Export it as a new photo or image DONE
- jpg's won't allow filters.
- returns null on jpg images
Copyright © 2020 FIXME
This program and the accompanying materials are made available under the terms of the Eclipse Public License 2.0 which is available at http://www.eclipse.org/legal/epl-2.0.
This Source Code may also be made available under the following Secondary Licenses when the conditions for such availability set forth in the Eclipse Public License, v. 2.0 are satisfied: GNU General Public License as published by the Free Software Foundation, either version 2 of the License, or (at your option) any later version, with the GNU Classpath Exception which is available at https://www.gnu.org/software/classpath/license.html.